Your website crashes due to traffic peaks? Use a cloud infrastructure!

What are the advantages of supporting your website with a scalable cloud infrastructure, which avoids server overload due to traffic peaks.


What measures are needed for a network "resilience" in times of Internet over-use? Many websites cannot keep up with the many connection requests, and their servers end up overloaded. 

However, a peak in online or e-commerce traffic should not be a problem, but a huge opportunity to acquire new customers. To make it possible, the IT must handle it appropriately so that it doesn’t impact response times. Just like the people queuing at the post office or supermarket, the overcrowding on your website should be spread over several counters or checkouts.

This can be done by relying on software with a cloud architecture, which is elastic and scalable. Compared to traditional infrastructure physically installed on corporate servers, this virtual infrastructure can modify its resources according to the actual demand. If traffic increases or decreases, the infrastructure will add or remove resources automatically.

What are the advantages of entrusting the delivery of content to software with this type of architecture?

1) Scalability and high performance

The website no longer relies on a single server, but on a virtual, cloud network that consists of several nodes, i.e. independent servers organized in clusters.

When the server is overloaded due to the many users connected, and is no longer able to meet the needs of the website, the system will create an instance, or a temporary virtual server, on a specific node, or will move it to freer nodes.

Let's take the example of an e-commerce site that in normal conditions requires four active virtual servers. On certain occasions (e.g. Black Friday) the traffic grows, so the infrastructure will make other virtual servers available and distribute the traffic over them (load balancing). As the peak decreases, the additional virtual servers are removed.

This ability is called "auto-scaling", meaning that the architecture automatically scales its resources to handle peak loads. As traffic is always distributed on free network nodes, quick response times are guaranteed.

The delivery of content on the website can also be accelerated through a CDN (Content Delivery Network) service, a network of servers that are geographically distributed worldwide. The result is excellent performance in terms of loading times.

Imagine we have a site of our company in Germany that also has contacts in Japan. Having a CDN will improve the performance of the site because a request from a user in Japan will be processed by the node closest to them. Latency, i.e. the time lapse between sending a data package and receiving it, is therefore reduced and the site experience is faster.

In addition, each node memorizes the data that is requested most frequently: in this way, it can be delivered to the user without having to request it from the source all over again.

2) High reliability and high uptime

High reliability is in the DNA of a cloud infrastructure. Because it is based on a multi-node architecture, whenever there are problems on a server, services will automatically migrate to the next available server and continue working.

With data replicated and accessible without interruption, there is a high throughput value, or transmission bandwidth. This means that the continuity of service is always automatically pursued, even in the event of hardware failures or software crashes.

The term uptime indicates the level of reliability and stability of the system. It is an essential factor in the choice of cloud computing services: the higher the uptime, the more reliable the service is.

THRON, for example, is a Saas (Software-as-a-Service) with dedicated cloud services provided by AWS (Amazon Web Services). It offers its users an online dashboard, certified by third parties, which provides the real-time status of all services and SLA (Service-Level-Agreement), and a monthly report on the service level including the average uptime (in the last 12 months it was 100%).

3) Security and full outsourcing

We have explained that a high reliability service is uninterrupted: in case of failure the user will use, without even realizing it, the next best working node.

But there is more. In order to ensure the highest levels of security and data integrity, the cloud can also provide disaster recovery and backup measures, which protect against any emergency.

Using this type of infrastructures also has an important economic advantage: full outsourcing. As the regulations regarding security, privacy and data processing continue to evolve, companies using such software are relieved of a large part of the management costs for updating and compliance of the systems, which remain in charge of the Saas provider.

There are other ways to verify the excellence of your Saas vendor. These include verifying whether or not they have ISO certification, a globally recognized and shared standard. ISO 27001 certified software, for example, has implemented all the best practices for a proper management of information security.

We can conclude that nowadays, with the existing technological solutions that can keep us safe from traffic peaks, congestion incidents deeply undermining user experience cannot and must not happen anymore. 

Cloud computing, and Saas in particular, provides us with a distributed and scalable infrastructure that automatically manages traffic peaks. It would be a shame not to take advantage of it, wouldn't it?